[Zope3-dev] Interesting parallels between Zope approach and N aked Objects

Martijn Faassen faassen@vet.uu.nl
Wed, 13 Nov 2002 23:27:20 +0100


Chris Withers wrote:
> Martijn Faassen wrote:
> >
> >People will have to think harder if they want to extend the framework,
> >as there's a bit more of a framework.
> 
> Then 'people' shouldn't have to do this all the time. The framework should 
> cover the majority of people's needs without the need to be extended.

Really? I disagree. I think the framework shouldn't do anything
people want. At least the utter minimum possible. This will make it
a lot more difficult to use, and harder to understand, but it will be
very worthwhile to have a framework that doesn't do much for people.
I will do my best to advocate this position on this mailing list.

Zope 3 shouldn't be a system that does what people want. To do something
they want, they would need to think hard each time they want 
something, as to do something they have to extend the framework 
each time. Frameworks are useful because of that; they're about having
to extend them all the time. This is why we are building a framework;
not about providing facilities, but because of the nice framework that
you should be building more of whenever you want to solve a problem.
People should be reinventing the wheel more often, and we should
make reinventing the wheel more easy (but still it should need
lots of brainpower to do so!).

This is the goal of building a framework. You seem to disagree and
seem to think we should create a framework that makes things *easier*.
That does common things for you, so you don't even have to reinvent the
wheel anymore. Next you'll ask for a user interface or something. Don't you
think zcml is good enough? Gosh, we might even have to pollute this
perfectly clean framework with mundane stuff like web page generation or
form handling or something..

> >Anyway, if you are concerned about people having to think too much why
> >don't you do something about it? 
> 
> Waving a flag is about all I have time for at the moment. I'd prefer to do 
> that than nothing at all and, from my point of view, see something I care 
> about head down an evolutionary blind alley.

I don't think you're being helpful by waving this particular flag. If you
cannot point out in more detail what you consider to be wrong, and if
you cannot offer more constructive suggestions, then please refrain 
from making these suggestions at all. Can you distinguish the highway from
the blind alley? Are you really suggesting we are building this framework
to make stuff harder for everybody? As that seems to be the inevitable
conclusion. Or is there anything specific you are worried about?
Oh, right, you don't have time to point out anything specific.

Regards,

Martijn